mnoukhov/nuevamol-135m-reinvent-sft
Partial fine-tune vs mnoukhov/nuevamol-135m-6B-wd3. Partial change: 59% of the weights changed (attention, embeddings, mlp, norms), in 30 of 30 layers.
Tensors changed: Embeddings 1/1, Norms 45/61, MLP 41/90, Attention 118/120